Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Intro to Social Hacking: How we lowered our cancellation rate by 90%

blog.joingrouper.com

Post reply on HN
Print view View original

21–30 of 43 posts

Re: Intro to Social Hacking: How we lowered our cancellation rate by 90%

#21

Post by waxman »

weff wrote: post #19

I've visited the website and I'm confused: are Facebook-less creatures unable to partake in the service? Wouldn't that be kind of weird?

Sorry, @weff.

Right now we require Facebook, because we heavily rely on the graph to match groups while requiring minimal work from our members. For instance, we want to make sure the two groups don't already know each other, and we haven't figured out a good way to replicate this functionality without Facebook (e.g. "tell us all the people you know!")

If you have any ideas, though, we'd love to hear them.

Re: Intro to Social Hacking: How we lowered our cancellation rate by 90%

#24

Post by LokiSnake »

waxman wrote: post #10
tkahn6 wrote: post #6

Cancellations through the service went down, but did actual cancellations (as in, the person just doesn't show) go down? If you make it harder to cancel you're relying even more heavily on the integrity or conscience of the cancelling party to notify the other group members of their desire to cancel. 90% seems like a really drastic decrease which is why I bring this up.

We call these "no-shows" and, fortunately our no-show rate is less than 0.2%. In other words, both groups show up more than 99.8% of the time. This will likely be the topic of a future blog post...

Just a hypothesis, but is this because users are committing in groups of 3? I bet no-shows would increase if you paired groups of 2 or 1.

Re: Intro to Social Hacking: How we lowered our cancellation rate by 90%

#25

Post by waxman »

itsprofitbaron wrote: post #20
waxman wrote: post #9

Earlier quoted context omitted.

Yep, it's definitely not their real a number. We built a simple Twilio app.

IMO, I'd highlight that its actually not their phone number rather its a Twilio number etc which is used to call them up. So I would change: we would give you a number for the other group, and YOU WOULD CALL THEM (not us) TO APOLOGIZE. To something like: we would give you a number for the other group which does not reveal their actual phone number, and YOU WOULD CALL THEM (not us) TO APOLOGIZE. The reason for this, i…

Fixed! Good call.

Re: Intro to Social Hacking: How we lowered our cancellation rate by 90%

#29

Post by EGreg »

I rarely upvote things to boomark them, but this is exactly the kind of thing that I like to read about.

We are building a platform for groups of people come together, kind of like grouper does for 3 guys and 3 girls, but a different approach. And psychology is a huge part of it! Making people want to do what you want them to do, or not want to do what you don't want them to do, is an art. But it is also a science, and there should be some sort of knowledge base of psychological tactics, in social apps or otherwise.

Anyone know of anything like that?

Re: Intro to Social Hacking: How we lowered our cancellation rate by 90%

#30

Post by weff »

waxman wrote: post #21
weff wrote: post #19

I've visited the website and I'm confused: are Facebook-less creatures unable to partake in the service? Wouldn't that be kind of weird?

Sorry, @weff. Right now we require Facebook, because we heavily rely on the graph to match groups while requiring minimal work from our members. For instance, we want to make sure the two groups don't already know each other, and we haven't figured out a good way to replicate this functionality without Facebook (e.g. "tell us all the people you know!") If you have any ideas, though, we'd love to hear them.

Yeah, I did not foresee that; it's actually quite critical you use Facebook.

Anything requiring the users to confirm they don't know the name of the other party slows the process quite some and removes the laissez-faire element.

Post reply on HN